Mathcad P-elements linear versus nonlinear stress 2014-t6
-
Upload
julio-banks -
Category
Engineering
-
view
96 -
download
1
Transcript of Mathcad P-elements linear versus nonlinear stress 2014-t6
P-Elements Linear Versus Nonlinear Stress 2014-T6.xmcd
Nonlinear analytical solution based on linear maximum stress at a hole for 2014-T6Aluminum Alloy
ORIGIN 1≡
References1. MMPDS-03, 1 October of 2006 2. ESDU 76016
Geometry of the hole: P 0.4375− 0.1660− 0.0880− 0.0100− 0.4375( )T in⋅≡
Far field traction
σg 36.428 ksi⋅≡
Maximum Edge Distance: e P5
P3
− 0.525 in=≡
Minimum Edge Distance: c P3
P1
− 0.349 in=≡
Plate Width: W e c+ 0.875 in=:=
Hole Diameter: D P4
P2
− 0.156 in=≡
Hole Radius: rD
20.078 in=≡
Nominal-Stress Path, from a to b: a r 0.078 in=:=
b c 0.349 in=:=
Dimensionless Hole location: ψe
c1.504=:=
Dimensionless Edge short-edge Distance: λD
2 c⋅0.2232=:=
Hole-Location Eccentricity: δψ 1−
2 ψ 1+( )⋅
W⋅ 0.088 in=:=
Julio C. Banks, P.E. page 1 of 23
P-Elements Linear Versus Nonlinear Stress 2014-T6.xmcd
Stress-strain Curve for 2014-T6 Aluminum Extrusions for t 0.500 in⋅≥
Fundamental Properties at room temperature per Reference 1, page 3-50
Modulus of Elasticity: E 10.8 103⋅ ksi⋅≡
Ultimate Tensile Stress: Ftu 64 ksi⋅≡
Yield Tensile Stress: Fty 58 ksi⋅≡
Ultimate Elongation: eu 7 %⋅≡
Poisson's Ratio: μ 0.33:= Only required for FEM solution
n = f(T): n 26≡ Reference 1, page 3-69
The following parameters are calculated in Appendix B
MMPDS Proportional Limit: SPL 53.08 ksi⋅= εPL 0.512 %⋅=
SPL
S0.291.5 %⋅=
StressCheck Required: S70E 58.327 ksi⋅= ε70E 0.772 %⋅=
Notice that the ultimate strain, eu, is measured after the test specimen has failed and there
is no load. Therefore, the actual ultimate strain at the ultimate load, Ftu 64 ksi⋅= , is:
εu 7.59 %⋅=
Julio C. Banks, P.E. page 2 of 23
P-Elements Linear Versus Nonlinear Stress 2014-T6.xmcd
Figure 1: MMPDS-03 Stress Strain for Aluminum Alloy Extrusion 2014-T6
Julio C. Banks, P.E. page 3 of 23
P-Elements Linear Versus Nonlinear Stress 2014-T6.xmcd
Perform a FEM study using the P-element formulation used in Stress Check forψ 1.504= and the Hole-Location Eccentricity δ 0.088 in=
Figure 2: Material parameters StressCheck requires to generate the Ramberg-OsgoodStress-strain data set
Julio C. Banks, P.E. page 4 of 23
P-Elements Linear Versus Nonlinear Stress 2014-T6.xmcd
Figure 3: Linear stress solution using P-element StressCheck FEA Software
Julio C. Banks, P.E. page 5 of 23
P-Elements Linear Versus Nonlinear Stress 2014-T6.xmcd
Figure 4: Nonlinear stress solution using P-element StressCheck FEA Software
Julio C. Banks, P.E. page 6 of 23
P-Elements Linear Versus Nonlinear Stress 2014-T6.xmcd
FEM from StressCheck
Linear Solution Nonlinear Solution
Maximum stress: σMax max σFEM Y( )( ) 116 ksi⋅=:= σ'Max max σ'FEM Y( )( ) 61.66 ksi⋅=:=
Short-edge stress: σNoma
b
yσFEM y( )⌠⌡
d
b a−44.59 ksi⋅=:= σ'Nom
a
b
yσ'FEM y( )⌠⌡
d
b a−44.05 ksi⋅=:=
Ultimate Stress: σu 1.50σNom⋅ 66.88 ksi⋅=:= σ'u 1.50σ'Nom⋅ 66.08 ksi⋅=:=
Gross SCF: Ktg_FEM
σMax
σg3.184=:= K'tg_FEM
σ'Max
σg1.693=:=
Nominal SCF: Ktn_FEM
σMax
σNom2.602=:= K'tn_FEM
σ'Max
σ'Nom1.400=:=
Nominal to gross transformation factor:
ΛFEM
Ktg_FEM
Ktn_FEM1.2240=:= Λ'FEM
K'tg_FEM
K'tn_FEM1.2093=:=
ΛFEM
σNom
σg1.2240=:= Λ'FEM
σ'Nom
σg1.2093=:=
MSu
Ftu
σu1− 0.04−=:= MS'u
Ftu
σ'u1− 0.03−=:=
Closed-form Solution
λ 0.2232= K'tg λ ψ, ( ) 3.184= σMax_CF K'tg λ ψ, ( ) σg⋅ 116.0 ksi⋅=:=
ψ 1.504= K'tn λ ψ, ( ) 2.481= ΛCF
K'tg λ ψ, ( )
K'tn λ ψ, ( )1.2830=:= or Λ λ ψ, ( ) 1.276=
The maximum strain found in Figure 5 is to be multiplied by 1.5 to obtain the ultimate strain toallow the calculation of the Margin of Safety based on strain rather than stress as follows
StressCheck FEM Maximum Nonlinear Strain: εMax_SC 1.757 %⋅:=
StressCheck Margin of Safety based on Strain: MSε
eu
1.5εMax_SC⋅1− 1.66=:=
Julio C. Banks, P.E. page 7 of 23
P-Elements Linear Versus Nonlinear Stress 2014-T6.xmcd
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.430
40
50
60
70
80
90
100
110
120
Linear VonMises Stress, ksiNolinear VonMises Stress, ksi
Von Mises Stress Vs Lateral Location in a Plate with Eccentric Hole
σ'Nom
ksi
σ'u
ksi
σFEM Y( )
ksi
σ'FEM Y( )
ksi
r
in
Y
in
Julio C. Banks, P.E. page 8 of 23
P-Elements Linear Versus Nonlinear Stress 2014-T6.xmcd
Determine a nonlinear Margin of Safety based upon the maximum strain at the hole dueto the negative Margin of Safety based on stress.
Figure 5: Nonlinear strain solution using P-element StressCheck FEA Software
Julio C. Banks, P.E. page 9 of 23
P-Elements Linear Versus Nonlinear Stress 2014-T6.xmcd
0.0 0.3 0.5 0.8 1.0 1.3 1.5 1.8 2.0 2.3 2.5 2.8 3.0 3.3 3.50.0
10.0
20.0
30.0
40.0
50.0
60.0
70.0
80.0
Ramberg-Osgood EquationRamberg-Osgood Equation Generated by StressCheck70% E Curve
Ramberg-Osgood Equation for Aluminum 2014-T6
Strain, %
Str
ess,
ksi
SPL
ksi
Ftu
ksi
εPL
%
ε70E
%
MMPDS 03 StressCheck Relative Error
SPL 53.084 ksi⋅= S'PL σSC εPL( ) 50.755 ksi⋅=:= %∆ SPL S'PL, ( ) 4.4− %⋅=
S70E 58.327 ksi⋅= S'70E σSC ε70E( ) 57.918 ksi⋅=:= %∆ S70E S'70E, ( ) 0.7− %⋅=
Su_RO 70.774 ksi⋅= S'u_RO σSC εu( ) 66.491 ksi⋅=:= %∆ Su_RO S'u_RO, ( ) 6.1− %⋅=
Note εu 7.593 %⋅= is the strain obtained from Ramberg-Osgood equation Corresponding to the loaded
test specimen. The actual ultimate strain provided in MMPDS (and similar reference books) is which is a measured strain of the ruptured (unloaded) test specimen
Julio C. Banks, P.E. page 10 of 23
P-Elements Linear Versus Nonlinear Stress 2014-T6.xmcd
Apply the ESED (Equivalent Strain-Energy Density also known as Glinka) Method to obtain thepseudo nonlinear solution using the linear FEM maximum stress at the short-edge of a hole in a plate in a tensile stress-field.
Strain-energy constant: Γ1
2εL⋅ σFEM⋅= 1a( )
σFEM E0 εL⋅= 2( )
Solve for the linear strain, εL from (2) as
εL
σFEM
E0= 3( )
Substitute (3) into (1) to get Γ1
2
σFEM2
E0⋅= 1b( )
Determine the strain at which the Ramberg-Osgood equates the strain-energy constant, Γ (Equation 1b)
The linear FEM Maximum Stress Using P-elements (StressCheck) is σFEM 116 ksi⋅≡
The strain-energy for the Given σFEM is Γ1
2
σFEM2
E⋅ 622.96 psi=:=
Initial guess: ε ε0.2 0.737 %⋅=:=
Given
Γ
0
ε
εσRO_rootε( )⌠⌡
d=
εG Find ε( ):=
The Pseudo Non-linear Strain is εG 1.289 %⋅=
Julio C. Banks, P.E. page 11 of 23
P-Elements Linear Versus Nonlinear Stress 2014-T6.xmcd
Post-process
σGlinka σRO_rootεG( ) 65.466 ksi⋅=:=The Pseudo Non-linear Strain is
The P-element (StressCheck) Maximum Non-linear Strain is εMax_SC 1.757 %⋅=
The ESED (Glinka) Maximum Pseudo Non-linear Strain is εG 1.289 %⋅=
Relative Error of the ESED strain solution with respect to the P-element (FEM) solution is
%∆ εMax_SC εG, ( ) 26.66− %⋅=
StressCheck Margin of Safety based on Strain: MSε
eu
1.5εMax_SC⋅1− 1.66=:=
ESED Margin of Safety based on Strain: MSu
eu
1.50εG⋅1− 2.62=:=
The P-element (StressCheck) Maximum Non-linear Stress is σ'Max 61.66 ksi⋅=
The ESED (Glinka) Maximum Pseudo Non-linear Stress is σGlinka 65.466 ksi⋅=
Relative Error of the ESED stress solution with respect to the P-element (FEM) solution is
%∆ σ'Max σGlinka, ( ) 6.17 %⋅=
ConclusionThe ESED (Equivalent-strain Energy Density) also known as Glinka can be used to obtain anestimate of the plastic strain and corresponding MS (Margin of Safety) when the linear solutionproduces a negative MS indicating nonlinear material behavior due to high loads.
It should be noted that this approach will most likely occur for the ultimate load Marginof Safety, MSu, rather than the limit load (yield) margin of safety, MSY
Julio C. Banks, P.E. page 12 of 23
P-Elements Linear Versus Nonlinear Stress 2014-T6.xmcd
Appendix A: StressCheck P-element FEM Solution Data and Cubic Spline Procedure
Ramberg-Osgood Numerical Results from StressCheck
Linear Solution Data Nonlinear Solution Data
DataSC
0.0000000
0.0043478
0.0086957
0.0130430
0.0173910
0.0217390
0.0260870
0.0304350
0.0347830
0.0391300
0.0434780
0.0478260
0.0521740
0.0565220
0.0608700
0.0652170
0.0695650
0.0739130
0.0782610
0.0826090
0.0869570
0.0913040
0.0956520
0.1000000
0
46872
59076
60992
62066
62818
63397
63870
64269
64615
64920
65193
65440
65666
65875
66068
66248
66416
66575
66724
66866
67001
67129
67251
≡
Data
0.1660−
0.1725−
0.1789−
0.1854−
0.1919−
0.1983−
0.2048−
0.2113−
0.2177−
0.2242−
0.2306−
0.2371−
0.2436−
0.2500−
0.2565−
0.2630−
0.2694−
0.2759−
0.2824−
0.2888−
0.2953−
0.3018−
0.3244−
0.3470−
0.3696−
0.3923−
0.4149−
0.4375−
116000.0
93230.0
78340.0
68310.0
61340.0
56320.0
52600.0
49750.0
47540.0
45780.0
44380.0
43270.0
42390.0
41700.0
41170.0
40760.0
40430.0
40160.0
39920.0
39690.0
39470.0
39265.0
38910.0
38450.0
37990.0
37370.0
36260.0
34270.0
≡ Data'
0.1660−
0.1725−
0.1789−
0.1854−
0.1919−
0.1983−
0.2048−
0.2113−
0.2177−
0.2242−
0.2306−
0.2371−
0.2436−
0.2500−
0.2565−
0.2630−
0.2694−
0.2759−
0.2824−
0.2888−
0.2953−
0.3018−
0.3141−
0.3264−
0.3388−
0.3511−
0.3635−
0.3758−
0.3881−
0.4005−
0.4128−
0.4252−
0.4375−
61660
61150
60580
59930
59180
58300
57250
55940
54230
51950
49330
47070
45430
44310
43550
42990
42510
42010
41480
40920
40370
39935
39790
39540
39270
39000
38740
38440
38040
37480
36660
35510
33920
≡
Julio C. Banks, P.E. page 13 of 23
P-Elements Linear Versus Nonlinear Stress 2014-T6.xmcd
FCS x X, Y, ( ) "Cubic-spline"
S cspline X Y, ( )←
y interp S X, Y, x, ( )←
≡
StressCheck Calculated Ramberg-Osgood Material Stress-strain Data
εSC_RO_Data DataSC1⟨ ⟩≡
σSC_RO_Data DataSC2⟨ ⟩ psi⋅≡
σSC ε( ) FCS ε εSC_RO_Data, σSC_RO_Data, ( )≡
StressCheck Linear solution P-elements StressCheckNonlinear solution P-elements
Y P3
Data1⟨ ⟩ in⋅−
≡ Y' P
3Data'1
⟨ ⟩in⋅−
≡
SSC Data2⟨ ⟩ psi⋅≡ S'SC Data'2⟨ ⟩
psi⋅≡
σFEM y( ) FCS y Y, SSC, ( )≡ σ'FEM y( ) FCS y Y', S'SC, ( )≡
Julio C. Banks, P.E. page 14 of 23
P-Elements Linear Versus Nonlinear Stress 2014-T6.xmcd
Appendix B: Functions
Ramberg-Osgood Stress-strain equation:
The defined 0.2% offset strain is: e0.2 0.2 %⋅≡
Yield Stress: S0.2 Fty 58 ksi⋅=≡
Yield Strain: ε0.2 e0.2
S0.2
E+ 0.74 %⋅=≡
0.2% offset Linear-stress: σL ε( ) E ε e0.2−( )⋅≡
Normalization stress: σn S0.2 58 ksi⋅=≡
Equation [9.8.4.1.2(b)]Ref. 1, page 9-200εRO σ( )
σ
Ee0.2
σ
σn
n
⋅+≡
Reference 1, Section 9.8.4.6.1 (page 9-200) states:If the proportional limit stress is equated with a plastic strain level of 0.0002 or a 0.02 percentdeviation from linearity, and the Ramberg- Osgood relationship is found to be valid for smallplastic strains, then the proportional limit stress, SPL fpl= , can be approximated from Equation
9.8.4.6(a) as follows:
Section [9.8.4.6.1]Ref. 1, page 9-200
Proportional Limit: SPL S0.2 0.1( )
1
n⋅ 53.08 ksi⋅=≡
εPL εRO SPL( ) 0.512 %⋅=≡
MMPDS 70%E Data point
S70E
S0.2
ksi
7
3e0.2⋅
E
ksi⋅
1
n
n
n 1−( )
ksi⋅
58.327 ksi⋅=≡ε70E 150
S0.2
70 %⋅ E⋅
n
⋅
1
n 1−( )
0.772 %⋅=≡
Define the 0.70xE straight line which when intercepting the Stress-strain curve will produce the S70Estress level required by StressCheck as σ70E ε( ) 0.70 E⋅( ) ε⋅≡
Ultimate Stress: Su Ftu 64 ksi⋅=≡ εu eu
Ftu
E+ 7.59 %⋅=≡
Relative-Error: %∆ a b, ( )b a−
a
≡
Julio C. Banks, P.E. page 15 of 23
P-Elements Linear Versus Nonlinear Stress 2014-T6.xmcd
The Ramberg-Osgood Equation [9.8.4.1.2(b)] is an implicit equation which needs to be solved using aroot-finding procedure since ε =f(σ), and not σ = g(ε). However, there is an approximation, σARO ε( ),
given in ESDU 76016, page 76016 as equation 5.1 which can be used as an estimate of the stress,σRO(ε), or even as a seed (starting value) for a more refined root-finding procedures.
σARO ε( ) "ESDU 76016, page 76016, equation 5.1"
k 0.79044 n⋅ 0.86977−←
β 11
n+
n k− 1−( )
11
n+
k−−←
λε E⋅σn
←
σEP ε E⋅ 1 β λk⋅+
1
nλ
n 1−⋅+
1−
n
⋅←
≡
Et σ( )E
1 e0.2 n⋅E
σn⋅
σ
σn
n
⋅+
≡ [9.8.4.2(b)]
Ref. 1, page 9-205
Et S0.2( ) 1.011 103× ksi⋅=
σRO_rootεRoot( ) "Root-finding Procedure by Julio C. Banks, P.E."
"Using Newton's rather than MathCAD's Method"
ε1
εRoot←
σ1
σARO ε1( )←
∆σTol 0.5 psi⋅←
∆σ 2 ∆σTol⋅←
i 1←
i i 1+←
εi
εRO σi 1−( )←
σi
σi 1− ε
iε
i 1−−( ) Et σi 1−( )⋅−←
∆σ σi
σi 1−−←
∆σ ∆σTol>while
σ σi
←
≡
Su_RO σRO_rootεu( ) 70.774 ksi⋅=≡
Julio C. Banks, P.E. page 16 of 23
P-Elements Linear Versus Nonlinear Stress 2014-T6.xmcd
Create a procedure that generates a set of data points corresponding to the Ramberg-Osgood Equation(1) call such a set of discrete Data Points. Assume a linear stress up until the Proportional Limit.
RO_Points NPe 2←
ε1
0 %⋅←
σ1
0 ksi⋅←
ε2
εPL←
σ2
SPL←
∆σ 0.8 ksi⋅←
NPp roundFtu SPL−
∆σ
←
NPep NPe NPp+ 2−←
σi
σi 1− ∆σ+←
εi
εRO σi( )←
i 3 NPep( )..∈for
σNPep 1+ Ftu←
εNPep 1+ εRO Ftu( )←
Data1⟨ ⟩ε←
Data2⟨ ⟩ σ
ksi←
Data
≡
Ramberg-Osgood Discrete set of Data Points
ε RO_Points1⟨ ⟩
≡ σRO RO_Points2⟨ ⟩
ksi⋅≡
Number of points generated, NRO length ε( ) 15=≡
The Ultimate Ramberg-Osgood Strains is εu_RO εNRO
3.18 %⋅=≡
Julio C. Banks, P.E. page 17 of 23
P-Elements Linear Versus Nonlinear Stress 2014-T6.xmcd
Appendix C: Determine the explicit equations for of the stress level corresponding to the interceptionpoint of the 0.70E line with the Ramberg-Osgood Curve
σ70E ε( ) 0.70 E⋅( ) ε⋅=C1( )
Ramberg-Osgood: εσ
Ee0.2
σ
S0.2
n
⋅+= Equation [9.8.4.1.2(b)]Ref. 1, page 9-200Renamed here as (C2)
Where n 1>
e0.2 0.20 %⋅=
S0.2 Fty=
Let the interception point be
ε ε0.70E=
S0.70E σ70E ε0.70E( )=
Introduce the interception point definition into equations C1 and C2 while simultaneouslyequating the strain from C1 with that of the Ramberg-Osgood equation C2
S0.70E
0.70 E⋅( )
S0.70E
Ee0.2
S0.70E
S0.2
n
⋅+=
Julio C. Banks, P.E. page 18 of 23
P-Elements Linear Versus Nonlinear Stress 2014-T6.xmcd
S0.70E
0.70 E⋅( )
S0.70E
Ee0.2
S0.70E
S0.2
n
⋅+=
1
0.701−
S0.70E
E⋅ e0.2
S0.70E
S0.2
n
⋅=
3
7
S0.70E
E⋅ e0.2
S0.70E
S0.2
n
⋅=
Since n 1> , then divide through by S0.70E( ) e0.2( )⋅
3
7 e0.2⋅
1
E⋅
S0.70E( ) n 1−( )
S0.2n
=
Solve for S0.70E : S0.70E
S0.2
7 e0.2⋅
3
E⋅
1
n
n
n 1−
=
Recall that e0.2 0.2 %⋅= therefore, 7 e0.2⋅ 0.014= , that is
S0.70E
S0.2
0.014
3
E⋅
1
n
n
n 1−
=3( )
Julio C. Banks, P.E. page 19 of 23
P-Elements Linear Versus Nonlinear Stress 2014-T6.xmcd
Determine the strain, at which the 0.70xE line, C1, intercepts the Ramberg-Osgood line, C2
ε70E
0.70 E⋅( ) ε70E⋅
Ee0.2
0.70 E⋅( ) ε70E⋅
S0.2
n
⋅+=
0.30ε70E⋅ e0.2
0.70 E⋅( ) ε70E⋅
S0.2
n
⋅=
Divide through by ε70E( ) e0.2( )⋅ 0.30
e0.2
0.70 E⋅S0.2
n
ε70En 1−( )⋅=
Solve for ε70E ε70E0.30
e0.2
S0.2
0.70 E⋅
n
⋅
1
n 1−
=
Again, e0.2 0.2 %⋅= therefore, 0.30
e0.2150= , that is
ε70E 150S0.2
0.70 E⋅
n
⋅
1
n 1−
= 4( )
The procedure for obtaining the explicit S0.70E from a cubic-spline stress-strain data set, or a
stress-strain model such as equation C2 (Ramberg-Osgood) is as follows:
Step 1: Solve ε70E for the given the yield strength, S0.2 Fty= , the modulus of elasticity, E, and
the exponent, n
Step 2: S0.70E FG ε70E( )=
Where the subscript of the function may be either RO, for Ramberg-Osgood, or CS for Cubic-spline interpolation function
It should be noted that if step 1 is not followed, then a root-finding procedure must be utilizedto numerically find the interception point, ε ε70E=
Julio C. Banks, P.E. page 20 of 23
P-Elements Linear Versus Nonlinear Stress 2014-T6.xmcd
Appendix D: Stress Concentration Factor (SCF) of a plate in tension with an eccentric hole.
Figure 1: Eccentric circular hole in finite-width plane Figure 2: Axial tension
Yield-stress Margin of Safety: MSy
Fty
σnom1−=
Ultimate-stress Margin of Safety: MSy
Ftu
1.50σnom⋅1−= or MSy
Ftu
σult1−=
Where σnom Λ λ ψ, ( ) σg⋅=B
A
yσ y( )⌠⌡
d
A B−= λ
d
2 c⋅= and ψ
e
c=
Peterson's Function (Reference 1): α 3.000 3.140− 3.667 1.527−( )T:=
Kn λ( ) α1
2
4
i
αiλ
i 1−⋅
∑
=
+:=1( )
2( )Λ λ ψ, ( )
1 λ2−
1 λ−( )
1
11
ψ1 1 λ
2−−( )⋅−
⋅≡
Kt λ ψ, ( ) Λ λ ψ, ( ) K'tn λ ψ, ( )⋅= 3( )
Equation (3) is implied in Chart 4.3 or reference 3.
Julio C. Banks, P.E. page 21 of 23
P-Elements Linear Versus Nonlinear Stress 2014-T6.xmcd
Chart 4.3 from Reference 3
References
1. Walter D. Pilkey, "Formulas for Stress, Strain and Structural Matrices", Second Edition. ISBN 0-471-03221-2, Pages 285, and 330 Through 332.2. Rudolph Earl Peterson, "Stress Concentration Factors", First Edition. ISBN 9780471683292, Page 152.3. Walter D. Pilkey and Deborah F. Pilkey "Peterson's Stress Concentration Factors", Third Edition.
Julio C. Banks, P.E. page 22 of 23
P-Elements Linear Versus Nonlinear Stress 2014-T6.xmcd
ORIGIN 1= ψ'1
ψ=
C
e= ψ 1.504= ψ'
1
ψ0.665=:=
Gross SCF Function
cg ψ( )
2.99690.0090
ψ−
0.01338
ψ2
+
0.12170.5180
ψ+
0.5297
ψ2
−
0.55650.7215
ψ+
0.6153
ψ2
+
4.04826.0146
ψ+
3.9815
ψ2
−
≡ cg ψ( )
2.997
0.232
1.309
6.287
=
K'tg λ ψ, ( ) cg ψ( )1
cg ψ( )2λ⋅+ cg ψ( )
3λ
2⋅+ cg ψ( )4λ
3⋅+≡
Nominal SCF Function
cn ψ( )
2.986
2.825−
2.446
=cn ψ( )
2.9890.0064
ψ−
2.872−0.095
ψ+
2.3480.196
ψ+
≡ cn ψ( )
2.985
2.809−
2.478
=
K'tn λ ψ, ( ) cn ψ( )1
cn ψ( )2λ⋅+ cn ψ( )
3λ
2⋅+≡
Determine the relationship between Kt λ ψ, ( ) and K'tn λ ψ, ( )
σmax K'tn λ ψ, ( ) σNom⋅= 4( )
σNom Λ λ ψ, ( ) σg⋅= 5( )
Substitute equation (5) into (4):
σmax K'tn λ ψ, ( ) Λ λ ψ, ( ) σg⋅( )⋅=Therefore,
σmax Kt λ ψ, ( ) σg⋅= 6( )
Where Kt λ ψ, ( ) Λ λ ψ, ( ) K'tn λ ψ, ( )⋅=
Julio C. Banks, P.E. page 23 of 23